A keyboard instrument includes at least one processor that determines a first pattern of intonation to be applied to a first time segment of a voice data on the basis of a first user operation on a first operation element, causes a first singing voice for the first time segment to be digitally synthesized from the first segment data in accordance with the determined first pattern of intonation, determines a second pattern of intonation to be applied to the second time segment of the voice data on the basis of a second user operation on a second operation element, and causes a second singing voice for the second time segment to be digitally synthesized from the second segment data in accordance with the determined second pattern of intonation.
|
1. A keyboard instrument comprising:
a keyboard that includes a row of a plurality of keys;
a plurality of operation elements provided behind the row of the plurality of keys on an instrument casing, the plurality of operation elements including a first operation element associated with a first segment data for a first time segment of a voice data that is to be output, and a second operation element associated with a second segment data for a second time segment that immediately follows the first time segment of the voice data; and
at least one processor,
wherein the at least one processor:
determines a first pattern of intonation to be applied to the first time segment of the voice data on the basis of a first user operation on the first operation element,
causes a first voice for the first time segment to be digitally synthesized from the first segment data in accordance with the determined first pattern of intonation and causes the digitally synthesized first voice to output,
determines a second pattern of intonation to be applied to the second time segment of the voice data on the basis of a second user operation on the second operation element, and
causes a second voice for the second time segment to be digitally synthesized from the second segment data in accordance with the determined second pattern of intonation and causes the digitally synthesized second voice to output, and
wherein the at least one processor, when the number of data segments in the voice data is greater than the number of the plurality of operation elements, causes the first operation element to be reassigned to segment data of the voice data for a time segment that comes after the second time segment after the digitally synthesized first voice is output.
7. A method performed by at least one processor in a keyboard instrument that includes, in addition to the at least one processor, a keyboard that includes a row of a plurality of keys; and a plurality of operation elements provided behind the row of the plurality of keys on an instrument casing, the plurality of operation elements including a first operation element associated with a first segment data for a first time segment of a voice data that is to be output, and a second operation element associated with a second segment data for a second time segment that immediately follows the first time segment of the voice data, the method comprising, via the at least one processor,
determining a first pattern of intonation to be applied to the first time segment of the voice data on the basis of a first user operation on the first operation element,
causing a first voice for the first time segment to be digitally synthesized from the first segment data in accordance with the determined first pattern of intonation and causes the digitally synthesized first voice to output,
determining a second pattern of intonation to be applied to the second time segment of the voice data on the basis of a second user operation on the second operation element, and
causing a second voice for the second time segment to be digitally synthesized from the second segment data in accordance with the determined second pattern of intonation and causes the digitally synthesized second voice to output,
wherein when the number of data segments in the voice data is greater than the number of the plurality of operation elements, the first operation element is reassigned to segment data of the voice data for a time segment that comes after the second time segment after the digitally synthesized first voice is output.
2. The keyboard instrument according to
wherein the number of the plurality of operation elements is eight, and the voice data contains at least nine segment data respectively for nine successive time segments from the first time segment to a ninth time segment, and
wherein the at least one processor associates, at a given timing, the first segment data through eighth segment data of the voice data with the plurality of operation elements, respectively, and the at least one processor causes the first operation element to be reassigned to the segment data of the voice data for the ninth time segment that follows the segment data for the eighth time segment after the digitally synthesized first voice is output.
3. The keyboard instrument according to
4. The keyboard instrument according to
wherein the plurality of operation elements are sliding operation elements, and
wherein, for each of the sliding operation elements, the at least one processor determines an intonation pattern from among a plurality of preset intonation patterns in accordance with an amount of slider operation on the sliding operation element.
5. The keyboard instrument according to
6. The keyboard instrument according to
wherein the at least one processor:
causes the trained acoustic model to output the acoustic feature data in response to the received lyric data and the received pitch data, and
digitally synthesizes an inferred voice that infers a voice of the singer on the basis of the acoustic feature data output by the trained acoustic model, and
causes the determined first pattern of intonation to be applied to the inferred voice in the first time segment and outputs the inferred voice that has been applied with the first pattern of intonation.
8. The method according to
wherein the number of the plurality of operation elements is eight, and the voice data contains at least nine segment data respectively for nine successive time segments from the first time segment to a ninth time segment, and
wherein at a given timing, the first segment data through eighth segment data of the voice data are associated with the plurality of operation elements, respectively, and the first operation element is reassigned to the segment data of the voice data for the ninth time segment that follows the segment data for the eighth time segment after the digitally synthesized first voice is output.
9. The method according to
10. The method according to
wherein the plurality of operation elements are sliding operation elements, and
wherein, for each of the sliding operation elements, an intonation pattern is determined from among a plurality of preset intonation patterns in accordance with an amount of slider operation on the sliding operation element.
11. The method according to
12. The method according to
wherein the keyboard instrument further includes a memory that stores a trained acoustic model obtained by performing machine learning on training musical score data including training lyric data and training pitch data, and on training voice data of a singer corresponding to the training musical score data, the trained acoustic model being configured to receive lyric data and pitch data and output acoustic feature data, and
wherein the method includes, via the at least one processor:
causing the trained acoustic model to output the acoustic feature data in response to the received lyric data and the received pitch data, and
digitally synthesizing an inferred voice that infers a voice of the singer on the basis of the acoustic feature data output by the trained acoustic model, and
causing the determined first pattern of intonation to be applied to the inferred voice in the first time segment and outputs the inferred voice that has been applied with the first pattern of intonation.
|
The present invention relates to a keyboard instrument, and a method performed by a computer in the keyboard instrument, with which the performance of rap or the like is possible.
There is a singing style known as “rap”. Rap is a musical technique in which spoken word or other such content is sung in time with the temporal progression of a musical rhythm, meter, or melody line. In rap, colorful musical expression is made possible by, among other things, the extemporaneous change of intonation.
Thus, as rap has both lyrics and flow (rhythm, meter, melody line), rap is extremely challenging to sing. If at least some of the musical elements in the aforementioned flow in rap were to be automated and the remaining musical elements able to be performed in time therewith using an electronic musical instrument or the like, rap would become accessible to even beginning rappers.
One known piece of conventional technology for automating singing is an electronic musical instrument that outputs a singing voice synthesized using concatenative synthesis, in which fragments of recorded speech are connected together and processed (for example, see Japanese Patent Application Laid-Open Publication No. H09-050287).
However, although with this conventional technology it is possible to specify pitch on the electronic musical instrument in time with the automatic progression of synthesized-voice-based singing, the intonation that is characteristic to rap cannot be controlled in real time.
Additionally, it has hitherto been difficult to apply sophisticated intonations in not only rap, but in other musical instrument performances as well.
An advantage of the present invention is that desired intonations are able to be applied in instrumental or vocal performances through a simple operation.
Additional or separate features and advantages of the invention will be set forth in the descriptions that follow and in part will be apparent from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
To achieve these and other advantages and in accordance with the purpose of the present invention, as embodied and broadly described, in one aspect, the present disclosure provides a keyboard instrument comprising: a keyboard that includes a row of a plurality of keys; a plurality of operation elements provided behind the row of the plurality of keys on an instrument casing, the plurality of operation elements including a first operation element associated with a first segment data for a first time segment of a voice data that is to be output, and a second operation element associated with a second segment data for a second time segment that immediately follows the first time segment of the voice data; and at least one processor, wherein the at least one processor: determines a first pattern of intonation to be applied to the first time segment of the voice data on the basis of a first user operation on the first operation element, causes a first voice for the first time segment to be digitally synthesized from the first segment data in accordance with the determined first pattern of intonation and causes the digitally synthesized first voice to output, determines a second pattern of intonation to be applied to the second time segment of the voice data on the basis of a second user operation on the second operation element, and causes a second voice for the second time segment to be digitally synthesized from the second segment data in accordance with the determined second pattern of intonation and causes the digitally synthesized second voice to output.
In another aspect, the present disclosure provides a method executed by the above-described at least one processor, including the above-enumerated processes performed by the at least one processor.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory, and are intended to provide further explanation of the invention as claimed.
Embodiments of the present invention will be described in detail below with reference to the drawings.
As illustrated in
The plurality of operation elements do not have to be sliding operation elements 105, and may be rotating operation elements (knob operation elements) 105 or button operation elements 105.
Using the RAM 203 as working memory, the CPU 201 executes an automatic performance control program stored in the ROM 202 and thereby controls the operation of the electronic keyboard instrument 100 in
The CPU 201 is provided with the timer 210 used in the present embodiment. The timer 210, for example, counts the progression of automatic performance in the electronic keyboard instrument 100.
Following a sound generation control instruction from the CPU 201, the sound source LSI 204 reads musical sound waveform data from a non-illustrated waveform ROM, for example, and outputs the musical sound waveform data to the D/A converter 211. The sound source LSI 204 is capable of 256-voice polyphony.
When the voice synthesis LSI 205 is given, as rap data 215, text data for lyrics and information relating to pitch by the CPU 201, the voice synthesis LSI 205 synthesizes voice data for a corresponding rap voice and outputs this voice data to the D/A converter 212.
The key scanner 206 regularly scans the pressed/released states of the keys on the keyboard 101 and the operation states of the switches on the first switch panel 102, the second switch panel 103, the bend sliders 105, and the bend switches 106 in
The LCD controller 208 is an integrated circuit (IC) that controls the display state of the LCD 104.
As illustrated in
Bend processor 320 is functionality whereby the CPU 201 in
The voice training section 301 and the voice synthesis section 302 in
Kei Hashimoto and Shinji Takaki, “Statistical parametric speech synthesis based on deep learning”, Journal of the Acoustical Society of Japan, vol. 73, no. 1 (2017), pp. 55-62
The voice training section 301 in
The voice training section 301, for example, uses voice sounds that were recorded when a given rap singer sang a plurality of rap songs as training rap voice data 312. Lyric text for each rap song is also prepared as training rap data 311.
The training text analysis unit 303 is input with training rap data 311, including lyric text, and the training text analysis unit 303 analyzes this data. The training text analysis unit 303 accordingly estimates and outputs a training linguistic feature sequence 313, which is a discrete numerical sequence expressing, inter alia, phonemes and pitches corresponding to the training rap data 311.
In addition to this input of training rap data 311, the training acoustic feature extraction unit 304 receives and analyzes training rap voice data 312 that was recorded via a microphone or the like when a given rap singer sang lyric text corresponding to the training rap data 311. The training acoustic feature extraction unit 304 accordingly extracts and outputs a training acoustic feature sequence 314 representing phonetic features corresponding to the training rap voice data 312.
The model training unit 305 uses machine learning to estimate an acoustic model with which the probability that a training acoustic feature sequence 314 will be generated given a training linguistic feature sequence 313 and an acoustic model is maximized. In other words, a relationship between a linguistic feature sequence (text) and an acoustic feature sequence (voice sounds) is expressed using a statistical model, which here is referred to as an acoustic model.
The model training unit 305 outputs, as training result 315, model parameters expressing the acoustic model that have been calculated through the employ of machine learning.
As illustrated in
The voice synthesis section 302, which is functionality performed by the voice synthesis LSI 205, includes a text analysis unit 307, the acoustic model unit 306, and a vocalization model unit 308. The voice synthesis section 302 performs statistical voice synthesis processing in which rap voice output data 217, corresponding to rap data 215 including lyric text, is synthesized by making predictions using a statistical model, which here is the acoustic model set in the acoustic model unit 306.
As a result of a performance by a performer made in concert with an automatic performance, the text analysis unit 307 is input with rap data 215, which includes information relating to phonemes, pitches, and the like for lyrics specified by the CPU 201 in
The acoustic model unit 306 is input with the linguistic feature sequence 316, and using this, the acoustic model unit 306 estimates and outputs an acoustic feature sequence 317 corresponding thereto. In other words, the acoustic model unit 306 estimates a value for an acoustic feature sequence 317 at which the probability that an acoustic feature sequence 317 will be generated based on a linguistic feature sequence 316 input from the text analysis unit 307 and an acoustic model set using the training result 315 of machine learning performed in the model training unit 305 is maximized.
The vocalization model unit 308 is input with the acoustic feature sequence 317. With this, the vocalization model unit 308 generates rap voice output data 217 corresponding to the rap data 215 including lyric text specified by the CPU 201. The rap voice output data 217 is output from the D/A converter 212, goes through the mixer 213 and the amplifier 214 in
The acoustic features expressed by the training acoustic feature sequence 314 and the acoustic feature sequence 317 include spectral information that models the vocal tract of a person, and sound source information that models the vocal cords of a person. A mel-cepstrum, line spectral pairs (LSP), or the like may be employed as spectral parameters. A power value and a fundamental frequency (F0) indicating the pitch frequency of the voice of a person may be employed as the sound source information. The vocalization model unit 308 includes a sound source generator 309 and a synthesis filter 310. The sound source generator 309 models the vocal cords of a person, and is sequentially input with a sound source information 319 sequence from the acoustic model unit 306. Thereby, the sound source generator 309, for example, generates a sound source signal that is made up of a pulse train (for voiced phonemes) that periodically repeats with a fundamental frequency (F0) and power value contained in the sound source information 319, that is made up of white noise (for unvoiced phonemes) with a power value contained in the sound source information 319, or that is made up of a signal in which a pulse train and white noise are mixed together. The synthesis filter 310 models the vocal tract of a person. The synthesis filter 310 forms a digital filter that models the vocal tract on the basis of a spectral information 318 sequence sequentially input thereto from the acoustic model unit 306, and using the sound source signal input from the sound source generator 309 as an excitation signal, generates and outputs rap voice output data 217 in the form of a digital signal.
The sampling frequency of the training rap voice data 312 is, for example, 16 kHz (kilohertz). When a mel-cepstrum parameter obtained through mel-cepstrum analysis, for example, is employed for a spectral parameter contained in the training acoustic feature sequence 314 and the acoustic feature sequence 317, the frame update period is, for example, 5 msec (milliseconds). In addition, when mel-cep strum analysis is performed, the length of the analysis window is 25 msec, and the window function is a twenty-fourth-order Blackman window function.
Next, a first embodiment of statistical voice synthesis processing performed by the voice training section 301 and the voice synthesis section 302 in
Shinji Sako, Keijiro Saino, Yoshihiko Nankaku, Keiichi Tokuda, and Tadashi Kitamura, “A trainable singing voice synthesis system capable of representing personal characteristics and singing styles”, Information Processing Society of Japan (IPSJ) Technical Report, Music and Computer (MUS) 2008 (12 (2008-MUS-074)), pp. 39-44, 2008 Feb. 8
In the first embodiment of statistical voice synthesis processing, when a user vocalizes lyrics in accordance with a given melody, HMM acoustic models are trained on how rap voice feature parameters, such as vibration of the vocal cords and vocal tract characteristics, change over time during vocalization. More specifically, the HMM acoustic models model, on a phoneme basis, spectrum and fundamental frequency (and the temporal structures thereof) obtained from the training rap data.
Next, a second embodiment of the statistical voice synthesis processing performed by the voice training section 301 and the voice synthesis section 302 in
Detailed description follows regarding operation for the automatic performance of songs, including rap, in embodiments of the electronic keyboard instrument 100 of
The specification of a bend curve and the application of a bend based thereon can be performed by a user in real time in a rap song that is progressing automatically using the volumes of the bend sliders 105 illustrated in
The bend switches 106, which function as a specification unit and are for example made up of 16 switches, are disposed above the bend sliders 105, which are for example made up of 16 sliders. Each switch of the bend switches 106 corresponds to the slider of the bend sliders 105 that is disposed directly therebelow. For any of the 16 beats, the user is able to disable the corresponding slider setting in the bend sliders 105 by turning OFF the corresponding switch in the bend switches 106. It is thereby possible to make it so that there is no bend effect on that beat.
The bend curve setting made for each of the 16 consecutive beats using the bend sliders 105 and the bend switches 106 is received by the bend processor 320 described in
Specifically, as each beat progresses, the bend processor 320 specifies, with respect to the voice synthesis section 302, pitch change information on the basis of the bend curve that is specified for that beat. The temporal resolution of pitch bends in one beat is, for example, 48. In this case, the bend processor 320 specifies, with respect to the voice synthesis section 302, and so pitch change information corresponding to the specified bend curve at timings obtained by dividing one beat by 48. The voice synthesis section 302 described in
In this manner, in the present embodiments, the lyrics and temporal progression, for example, of a rap song are left to be automatically performed, making it possible for the user to specify bend curve intonation patterns for rap-like pitches, for example, per each unit of progression (e.g., beat), and making it possible for the user to freely enjoy rap performances.
In particular, in this case, using the bend sliders 105 and bend switches 106 corresponding to each of, e.g., 16 beats, the user is able to specify, in real time, a bend curve for realizing a rap voice pitch at each beat per every 16 beats in an automatic performance that is progressing automatically, making it possible for the user to put on their own rap performance as the rap song is performed automatically.
The specification of a bend curve for each beat, for example, may be performed by a user in advance and stored in association with a rap song to be automatically performed such that when the rap song is automatically performed, the bend processor 320 loads the specified bend curves and designates, with respect to the voice synthesis section 302, intonations for the pitch of the rap voice corresponding to the bend curve that has been specified.
Thereby, users are able to apply intonation to the pitch of a rap voice in a rap song in a deliberate manner.
Incidentally, the number of segments in voice data (which encompasses various forms of data, such as musical piece data, lyric data, and text data) is typically greater than the number of the plurality of operation elements (sliding operation elements 105). For this reason, the processor 201 performs processing in which, after the output of first segment data that was associated with a first operation element, segment data associated with the first operation element is changed from first segment data to segment data that comes after the first segment data.
Suppose that the number of the plurality of operation elements (sliding operation elements 105) was equal to eight. In this case, the processor 201 would, at a given timing, associate the plurality of operation elements with, for example, segments of voice data that are two measures long. In other words, at a given timing, the plurality of operation elements are given associations as follows:
First operation element . . . first segment data (segment for the first beat in a first measure)
Second operation element . . . second segment data (segment for the second beat in the first measure)
Third operation element . . . third segment data (segment for the third beat in the first measure)
Fourth operation element . . . fourth segment data (segment for the fourth beat in the first measure)
Fifth operation element . . . fifth segment data (segment for the first beat in a second measure)
Sixth operation element . . . sixth segment data (segment for the second beat in the second measure)
Seventh operation element . . . seventh segment data (segment for the third beat in the second measure)
Eighth operation element . . . eighth segment data (segment for the fourth beat in the second measure)
After the keyboard instrument outputs first segment data that was associated with the first operation element, the processor 201 performs processing in which segment data associated with the first operation element is changed from first segment data to ninth segment data that follows the eighth segment data (for example, a segment for the first beat in a third measure).
In other words, during a performance, segment data allocated to a first operation element is successively changed in the manner: first segment data→ninth segment data→17th segment data, and so on. That is, for example, at a timing at which the production of a singing voice up to the fourth beat in the first measure ends, segment data allocated to the operation elements is as follows:
First operation element . . . ninth segment data (segment for the first beat in a third measure)
Second operation element . . . 10th segment data (segment for the second beat in the third measure)
Third operation element . . . 11th segment data (segment for the third beat in the third measure)
Fourth operation element . . . 12th segment data (segment for the fourth beat in the third measure)
Fifth operation element . . . 13th segment data (segment for the first beat in a fourth measure)
Sixth operation element . . . 14th segment data (segment for the second beat in the fourth measure)
Seventh operation element . . . 15th segment data (segment for the third beat in the fourth measure)
Eighth operation element . . . 16th segment data (segment for the fourth beat in the fourth measure)
An advantage of the present invention is that despite having only a limited number of operation elements, during a performance, because the segment of voice data allocated to a single operation element changes, the voice data is able to be sung in a satisfactory manner no matter what the length of the voice data.
Combinations of intonation patterns allocated to respective operation elements, for example, a combination of intonation patterns in which intonation pattern 401 (#0) (a first pattern) is allocated to the first operation element and intonation pattern 401 (#1) (a second pattern) is allocated to the second operation element, also do not change so long as the operation elements 105 are not operated. Accordingly, once a combination of intonation patterns has been determined by operation of the operation elements 105, even if the user does not subsequently operate the operation elements 105, the keyboard instrument is able to produce sound using the determined combination of intonation patterns from the start to the end of the voice data. In other words, during a performance in which the keyboard 101 is operated by the user, it is not necessary to operate the operation elements 105 to apply intonation to a singing voice. This has the advantage of enabling the user to concentrate on operation of the keyboard 101.
The combination of intonation patterns is of course able to be changed at any time in the middle of a performance if the user operates the operation elements 105. In other words, during a performance in which the keyboard 101 is operated by the user, combinations of intonation patterns can be changed in concert with changes in expression in the performance. This has the advantage of enabling the user to continue performing in an enjoyable manner.
In the example of
In the present example, a singing voice is synthesized on the basis of pitch data that has been specified through operation of the keyboard 101 by the user. In other words, singing voice data that corresponds to a lyric and a specified pitch is generated in real time.
The musical piece data is configured by data blocks called “chunks”. Specifically, the musical piece data is configured by a header chunk at the beginning of the file, a first track chunk that comes after the header chunk and stores lyric data for a lyric part, and a second track chunk that stores performance data for an accompaniment part.
The header chunk is made up of five values: ChunkID, ChunkSize, FormatType, NumberOfTrack, and TimeDivision. ChunkID is a four byte ASCII code “4D 54 68 64” (in base 16) corresponding to the four half-width characters “MThd”, which indicates that the chunk is a header chunk. ChunkSize is four bytes of data that indicate the length of the FormatType, NumberOfTrack, and TimeDivision part of the header chunk (excluding ChunkID and ChunkSize). This length is always “00 00 00 06” (in base 16), for six bytes. FormatType is two bytes of data “00 01” (in base 16). This indicates that in the case of the present embodiments, the format type is format 1, in which multiple tracks are used. NumberOfTrack is two bytes of data “00 02” (in base 16). This indicates that in the case of the present embodiments, two tracks, corresponding to the lyric part and the accompaniment part, are used. TimeDivision is data indicating a timebase value, which itself indicates resolution per quarter note. TimeDivision is two bytes of data “01 E0” (in base 16). In the case of the present embodiments, this indicates 480 in decimal notation.
The first and second track chunks are each made up of a ChunkID, ChunkSize, and performance data pairs. The performance data pairs are made up of DeltaTime_1[i] and Event_1[i] (for the first track chunk/lyric part), or DeltaTime_2[i] and Event_2[i] (for the second track chunk/accompaniment part). Note that 0≤i≤L for the first track chunk/lyric part, and 0≤i≤M for the second track chunk/accompaniment part. ChunkID is a four byte ASCII code “4D 54 72 6B” (in base 16) corresponding to the four half-width characters “MTrk”, which indicates that the chunk is a track chunk. ChunkSize is four bytes of data that indicate the length of the respective track chunk (excluding ChunkID and ChunkSize).
DeltaTime_1[i] is variable-length data of one to four bytes indicating a wait time (relative time) from the execution time of Event_1[i−1] immediately prior thereto. Similarly, DeltaTime_2[i] is variable-length data of one to four bytes indicating a wait time (relative time) from the execution time of Event_2[i−1] immediately prior thereto. Event_1[i] is a meta event designating the vocalization timing and pitch of a rap lyric in the first track chunk/lyric part. Event_2[i] is a MIDI event designating “note on” or “note off” or is a meta event designating time signature in the second track chunk/accompaniment part. In each DeltaTime_1[i] and Event_1[i] performance data pair of the first track chunk/lyric part, Event_1[i] is executed after a wait of DeltaTime_1[i] from the execution time of the Event_1[i−1] immediately prior thereto. The vocalization and progression of lyrics is realized thereby. In each DeltaTime_2[i] and Event_2[i] performance data pair of the second track chunk/accompaniment part, Event_2[i] is executed after a wait of DeltaTime_2[i] from the execution time of the Event_2[i−1] immediately prior thereto. The progression of automatic accompaniment is realized thereby.
After first performing initialization processing (step S801), the CPU 201 repeatedly performs the series of processes from step S802 to step S808.
In this repeat processing, the CPU 201 first performs switch processing (step S802). Here, based on an interrupt from the key scanner 206 in
Next, based on an interrupt from the key scanner 206 in
Next, the CPU 201 processes data that should be displayed on the LCD 104 in
Next, the CPU 201 performs rap playback processing (step S805). In this processing, the CPU 201 performs a control process described in
Then, the CPU 201 performs sound source processing (step S806). In the sound source processing, the CPU 201 performs control processing such as that for controlling the envelope of musical sounds being generated in the sound source LSI 204.
Finally, the CPU 201 determines whether or not a performer has pressed a non-illustrated power-off switch to turn off the power (step S807). If the determination of step S807 is NO, the CPU 201 returns to the processing of step S802. If the determination of step S807 is YES, the CPU 201 ends the control process illustrated in the flowchart of
First, in
TickTime(sec)=60/Tempo/TimeDivision (1)
Accordingly, in the initialization processing illustrated in the flowchart of
Next, the CPU 201 sets a timer interrupt for the timer 210 in
Bend processing, described later, is performed in units of time obtained by multiplying 1 TickTime by D. D is calculated according to Equation (2) below. This equation uses the timebase value TimeDivision indicating resolution per quarter note, described in
D=TimeDivision/R (2)
As in the foregoing, if, for example, each quarter note (one beat in the case of a 4/4 time signature) is equal to 480 TickTime and R=48, bend processing would be performed every D=480/R=480/48=10 TickTime.
Then, the CPU 201 performs additional initialization processing, such as that to initialize the RAM 203 in
The flowcharts in
First, the CPU 201 determines whether or not the tempo of lyric progression and automatic accompaniment has been changed using a switch for changing tempo on the first switch panel 102 in
Next, the CPU 201 determines whether or not a rap song has been selected with the second switch panel 103 in
Then, the CPU 201 determines whether or not a switch for starting a rap on the first switch panel 102 in
Then, the CPU 201 determines whether or not a bend-curve-setting start switch on the first switch panel 102 in
Finally, the CPU 201 determines whether or not any other switches on the first switch panel 102 or the second switch panel 103 in
Similarly to at step S901 in
Next, similarly to at step S902 in
First, with regards to the progression of an automatic performance, the CPU 201 initializes the value of an ElapseTime variable in the RAM 203 for indicating, in units of TickTime, the amount of time that has elapsed since the start of the automatic performance to 0. The CPU 201 also initializes the values of both a DeltaT_1 (first track chunk) variable and a DeltaT_2 (second track chunk) variable in the RAM 203 for counting, similarly in units of TickTime, relative time since the last event to 0. Next, the CPU 201 initializes the respective values of an AutoIndex_1 variable in the RAM 203 for specifying an i value (1≤i≤L−1) for DeltaTime_1[i] and Event_1[i] performance data pairs in the first track chunk of the musical piece data illustrated in
Next, the CPU 201 initializes the value of a SongIndex variable in the RAM 203, which designates the current rap position, to 0 (step S922).
The CPU 201 also initializes the value of a SongStart variable in the RAM 203, which indicates whether to advance (=1) or not advance (=0) the lyrics and accompaniment, to 1 (advance) (step S923).
Then, the CPU 201 determines whether or not a performer has configured the electronic keyboard instrument 100 to playback an accompaniment together with rap lyric playback using the first switch panel 102 in
If the determination of step S924 is YES, the CPU 201 sets the value of a Bansou variable in the RAM 203 to 1 (has accompaniment) (step S925). Conversely, if the determination of step S924 is NO, the CPU 201 sets the value of the Bansou variable to 0 (no accompaniment) (step S926). After the processing at step S925 or step S926, the CPU 201 ends the rap-starting processing at step S1006 in
Next, the CPU 201 acquires rap lyric data for the 16 beats (four measures worth) that were specified in step S1101 from the ROM 202 (step S1102). The CPU 201 can display rap lyric data acquired in this manner on the LCD 104 in
Next, the CPU 201 sets an initial value for a beat position in the 16 consecutive beats to 0 (step S1103).
Then, after initializing, to 0, the value of a variable i in the RAM 203 that indicates beat position in the 16 consecutive beats in step S1103, while incrementing the value of i by 1 at step S1106, the CPU 201 repeatedly performs step S1104 and step S1105 (for any of #0-#3) for the 16 beats until the value of i is determined to have exceeded 15 at step S1107.
In this repeat processing, the CPU 201 first loads a slider value (s) of the slider at beat position i in the bend sliders 105 described in
Next, if the slider value s at beat position i is equal to 0, the CPU 201 stores the number 0, for bend curve 401 (#0) in
Measure number=(measure number specified at S1101)+(the integer part of 4/i) (3)
Beat number=the remainder of beat position i/4 (4)
Next, if the slider value s at beat position i is equal to 1, the CPU 201 stores the number 1, for bend curve 401 (#1) in
Next, if the slider value s at beat position i is equal to 2, the CPU 201 stores the number 2, for bend curve 401 (#2) in
Next, if the slider value s at beat position i is equal to 3, the CPU 201 stores the number 3, for bend curve 401 (#3) in
When the value of the variable i is determined to have reached 15 at step S1107 in this repeat processing, the CPU 201 ends the processing of the flowchart in
First, the CPU 201 performs a series of processes corresponding to the first track chunk (steps S1201 to S1207). The CPU 201 starts by determining whether or not the value of SongStart is equal to 1, in other words, whether or not advancement of the lyrics and accompaniment has been instructed (step S1201).
When the CPU 201 has determined there to be no instruction to advance the lyrics and accompaniment (the determination of step S1201 is NO), the CPU 201 ends the automatic-performance interrupt processing illustrated in the flowchart of
When the CPU 201 has determined there to be an instruction to advance the lyrics and accompaniment (the determination of step S1201 is YES), the value of the ElapseTime variable in the RAM 203, which indicates the amount of time that has elapsed since the start of the automatic performance in units of TickTime, is incremented by 1. Because the automatic-performance interrupt processing of
Next, the CPU 201 then determines whether or not the value of DeltaT_1, which indicates the relative time since the last event in the first track chunk, matches the wait time DeltaTime_1[AutoIndex_1] of the performance data pair indicated by the value of AutoIndex_1 that is about to be executed (step S1203).
If the determination of step S1203 is NO, the CPU 201 increments the value of DeltaT_1, which indicates the relative time since the last event in the first track chunk, by 1, and the CPU 201 allows the time to advance by 1 TickTime corresponding to the current interrupt (step S1204). Following this, the CPU 201 proceeds to step S1208, which will be described later.
If the determination of step S1203 is YES, the CPU 201 executes the first track chunk event Event_1[AutoIndex_1] of the performance data pair indicated by the value of AutoIndex_1 (step S1205). This event is a rap event that includes lyric data.
Then, the CPU 201 stores the value of AutoIndex_1, which indicates the position of the rap event that should be performed next in the first track chunk, in the SongIndex variable in the RAM 203 (step S1205).
The CPU 201 then increments the value of AutoIndex_1 for referencing the performance data pairs in the first track chunk by 1 (step S1206).
Next, the CPU 201 resets the value of DeltaT_1, which indicates the relative time since the rap event most recently referenced in the first track chunk, to 0 (step S1207). Following this, the CPU 201 proceeds to the processing at step S1208.
Next, the CPU 201 performs a series of processes corresponding to the second track chunk (steps S1208 to S1214). The CPU 201 starts by determining whether or not the value of DeltaT_2, which indicates the relative time since the last event in the second track chunk, matches the wait time DeltaTime_2[AutoIndex_2] of the performance data pair indicated by the value of AutoIndex_2 that is about to be executed (step S1208).
If the determination of step S1208 is NO, the CPU 201 increments the value of DeltaT_2, which indicates the relative time since the last event in the second track chunk, by 1, and the CPU 201 allows the time to advance by 1 TickTime corresponding to the current interrupt (step S1209). Following this, the CPU 201 proceeds to the bend processing at step S1211.
If the determination of step S1208 is YES, the CPU 201 then determines whether or not the value of the Bansou variable in the RAM 203 that denotes accompaniment playback is equal to 1 (has accompaniment) (step S1210) (see steps S924 to S926 in
If the determination of step S1210 is YES, the CPU 201 executes the second track chunk accompaniment event Event_2[AutoIndex_2] indicated by the value of AutoIndex_2 (step S1211). If the event Event_2[AutoIndex_2] executed here is, for example, a “note on” event, the key number and velocity specified by this “note on” event are used to issue a command to the sound source LSI 204 in
However, if the determination of step S1210 is NO, the CPU 201 skips step S1211 and proceeds to the processing at the next step S1212 without executing the current accompaniment event Event_2[AutoIndex_2]. Here, in order to progress in sync with the lyrics, the CPU 201 performs only control processing that advances events.
After step S1211, or when the determination of step S1210 is NO, the CPU 201 increments the value of AutoIndex_2 for referencing the performance data pairs for accompaniment data in the second track chunk by 1 (step S1212).
Next, the CPU 201 resets the value of DeltaT_2, which indicates the relative time since the event most recently executed in the second track chunk, to 0 (step S1213).
Then, the CPU 201 determines whether or not the wait time DeltaTime_2[AutoIndex_2] of the performance data pair indicated by the value of AutoIndex_2 to be executed next in the second track chunk is equal to 0, or in other words, whether or not this event is to be executed at the same time as the current event (step S1214).
If the determination of step S1214 is NO, the CPU 201 proceeds to the bend processing of step S1211.
If the determination of step S1214 is YES, the CPU 201 returns to step S1210, and repeats the control processing relating to the event Event_2[AutoIndex_2] of the performance data pair indicated by the value of AutoIndex_2 to be executed next in the second track chunk. The CPU 201 repeatedly performs the processing of steps S1210 to S1214 the same number of times as there are events to be simultaneously executed. The above processing sequence is performed when a plurality of “note on” events are to generate sound at simultaneous timings, as for example happens in chords and the like.
After the processing at step S1209, or if the determination of step S1214 is NO, the CPU 201 performs bend processing (step S1211). Here, on the basis of the bend curve settings of each measure, and each beat in the measures, that have been set in the bend curve settings table 600 illustrated in
First, at step S1205 in the automatic-performance interrupt processing of
If the determination of step S1301 is YES, that is, if the present time is a rap playback timing, the CPU 201 then determines whether or not a new performer key press on the keyboard 101 in
If the determination of step S1302 is YES, the CPU 201 sets the pitch specified by a performer key press to a non-illustrated register, or to a variable in the RAM 203, as a vocalization pitch (step S1303).
Then, the CPU 201 reads the rap lyric string from the rap event Event_1[SongIndex] in the first track chunk of the musical piece data in the RAM 203 indicated by the SongIndex variable in the RAM 203. The CPU 201 generates rap data 215 for vocalizing, at the vocalization pitch set to the pitch based on a key press that was set at step S1303, rap voice output data 217 corresponding to the lyric string that was read, and instructs the voice synthesis LSI 205 to perform vocalization processing (step S1305). The voice synthesis LSI 205 performs the statistical voice synthesis processing described with reference to
If at step S1301 it is determined that the present time is a rap playback timing and the determination of step S1302 is NO, that is, if it is determined that no new key press is detected at the present time, the CPU 201 reads the data for a pitch from the rap event Event_1[SongIndex] in the first track chunk of the musical piece data in the RAM 203 indicated by the SongIndex variable in the RAM 203, and sets this pitch to a non-illustrated register, or to a variable in the RAM 203, as a vocalization pitch (step S1304).
In the case of a rap performance, pitch may, or may not be, linked with the pitch of a melody.
Then, by performing the processing at step S1305, described above, the CPU 201 generates rap data 215 for vocalizing, at the vocalization pitch set at step S1304, rap voice output data 217 corresponding to the lyric string that was read from the rap event Event_1[SongIndex], and instructs the voice synthesis LSI 205 to perform vocalization processing (step S1305). In performing the statistical voice synthesis processing described with reference to
After the processing of step S1305, the CPU 201 stores the rap position at which playback was performed indicated by the SongIndex variable in the RAM 203 in a SongIndex_pre variable in the RAM 203 (step S1306).
Then, the CPU 201 clears the value of the SongIndex variable so as to become a null value and makes subsequent timings non-rap playback timings (step S1307). The CPU 201 subsequently ends the rap playback processing at step S805 in
If the determination of step S1301 is NO, that is, if the present time is not a rap playback timing, the CPU 201 then determines whether or not a new performer key press on the keyboard 101 in
If the determination of step S1308 is NO, the CPU 201 ends the rap playback processing at step S805 in
If the determination of step S1308 is YES, the CPU 201 generates rap data 215 instructing that the pitch of the rap voice output data 217 currently undergoing vocalization processing in the voice synthesis LSI 205, which corresponds to the lyric string for rap event Event_1[SongIndex_pre] in the first track chunk of the musical piece data in the RAM 203 indicated by the SongIndex_pre variable in the RAM 203, is to be changed to the pitch based on the performer key press detected at step S1308, and outputs the rap data 215 to the voice synthesis LSI 205 (step S1309). At such time, the frame in the rap data 215 where a latter phoneme among phonemes in the lyrics already being subjected to vocalization processing starts is set as the starting point of the change to the specified pitch. For example, in the case of the lyric string “Ki”, the this is the frame where the latter phoneme /i/ in the constituent phoneme sequence /k/ /i/ starts. The voice synthesis LSI 205 performs the statistical voice synthesis processing described with reference to
Due to the processing at step S1309, the pitch of vocalization of rap voice output data 217 vocalized from an original timing immediately before the current key press timing is able to be changed to the pitch played by the performer and continue being vocalized at the current key press timing.
After the processing at step S1309, the CPU 201 ends the rap playback processing at step S805 in
Then, the CPU 201 determines whether or not the value of the DividingTime variable matches the value of D calculated using Equation (2) (step S1402). If the determination of step S1402 is NO, the CPU 201 ends the bend processing at step S1211 in
If the determination of step S1402 is YES, the CPU 201 resets the value of the DividingTime variable to 0 (step S1403).
Next, the CPU 201 determines whether or not the value of the BendAdressOffset variable in the RAM 203 matches the last address R−1 in one bend curve (step S1404). Here, the CPU 201 determines whether or not bend processing with respect to a single beat has ended. Because the value of the BendAdressOffset variable is initialized to R−1 in step S921 of the rap-starting processing of
If the determination of step S1404 is YES, the CPU 201 resets the value of the BendAdressOffset variable to 0, which indicates the beginning of a bend curve (see
Then, the CPU 201 calculates the current measure number and beat number from the value of the ElapseTime variable (step S1406). In the case of a 4/4 time signature, because the number of TickTimes per beat is given in terms of the value of TimeDivision, the ElapseTime variable is divided by the value of TimeDivision, and the result thereof is further divided by four (the number of beats per measure), whereby the current measure number and beat number can be calculated.
Next, the CPU 201 acquires the bend curve number corresponding to the measure number and beat number calculated at step S1406 from the bend curve settings table 600 illustrated in
However, if the value of the BendAdressOffset variable in the RAM 203 has not reached the last address R−1 in one bend curve and the determination of step S1404 is NO, the CPU 201 increments the value of the BendAdressOffset variable indicating the offset address in the bend curve by 1 (step S1409).
Next, the CPU 201 determines whether or not a bend curve number was assigned to CurveNum variable data by the processing of step S1407 in the current or previous automatic-performance interrupt processing (step S1408).
If the determination of step S1408 is YES, the CPU 201 adds the offset value assigned to the BendAdressOffset variable to the beginning address BendCurve[CurveNum] in the bend curve data in the ROM 202 corresponding to the bend curve number assigned to the CurveNum variable, and acquires a bend value from the resulting address in the bend curve table 700 (see
Finally, similarly to described for step S1309 in
If no bend curve number is assigned to the CurveNum variable and the determination of step S1408 is NO, because the bend curve setting has been disabled by the user for that beat, the CPU 201 ends the bend processing at step S1211 in
In this manner, in the present embodiment, bend processing corresponding to a bend curve that is specified in real time or has been specified in advance by a user for each beat is able to be performed with respect to rap sounds.
In addition to the embodiments described above, when the bend processor 320 in
In the embodiments described above, a user sets a bend curve per beat within, for example, 16 consecutive beats (four measures in the case of a 4/4 time signature). However, a user interface may be employed that specifies, en bloc, 16 beat bend curve sets. This makes it easy to make specifications that imitate rap performances by well-known rap singers.
A emphasis unit may also be provided that changes bend curves and emphasizes intonations either randomly or every given number of consecutive beats (e.g., four beats), such as at the beginning of a measure. This makes a greater variety of rap expressions possible.
In the embodiments above, bend processing is performed as a pitch bend of the pitch of a rap voice. However, bend processing may be performed with respect to aspects other than pitch, such as, for example, the intensity or tone color of sounds. This makes a greater variety of rap expressions possible.
In the embodiments above, the specification of intonation patterns is performed with respect to a rap voice. However, the specification of intonation patterns may be performed with respect to sounds other than of a rap voice, such as musical information for musical instrument sounds.
In the first embodiment of statistical voice synthesis processing employing HMM acoustic models described with reference to
In the second embodiment of statistical voice synthesis processing employing a DNN acoustic model described with reference to
In the embodiments described above, statistical voice synthesis processing techniques, employed as voice synthesis methods, can be implemented with markedly less memory capacity compared to conventional concatenative synthesis. For example, in an electronic musical instrument that uses concatenative synthesis, memory having several hundred megabytes of storage capacity is needed for voice sound fragment data. However, the present embodiments get by with memory having just a few megabytes of storage capacity in order to store training result 315 model parameters in
Moreover, with conventional fragmentary data methods, it takes a great deal of time (years) and effort to produce data for rap performances since fragmentary data needs to be adjusted by hand. However, because almost no data adjustment is necessary to produce training result 315 model parameters for the HMM acoustic models or the DNN acoustic model of the present embodiments, performance data can be produced with only a fraction of the time and effort. This also makes it possible to provide a lower cost electronic musical instrument. Further, using a server computer 300 available for use as a cloud service, or training functionality built into the voice synthesis LSI 205, general users can train the electronic musical instrument using their own voice, the voice of a family member, the voice of a famous person, or another voice, and have the electronic musical instrument give a rap performance using this voice for a model voice. In this case too, rap performances that are markedly more natural and have higher quality sound than hitherto are able to be realized with a lower cost electronic musical instrument.
In the embodiments described above, the present invention is embodied as an electronic keyboard instrument. However, the present invention can also be applied to electronic string instruments and other electronic musical instruments.
Voice synthesis methods able to be employed for the vocalization model unit 308 in
In the embodiments described above, a first embodiment of statistical voice synthesis processing in which HMM acoustic models are employed and a subsequent second embodiment of a voice synthesis method in which a DNN acoustic model is employed were described. However, the present invention is not limited thereto. Any voice synthesis method using statistical voice synthesis processing may be employed by the present invention, such as, for example, an acoustic model that combines HMMs and a DNN.
In the embodiments described above, rap lyric information is given as musical piece data. However, text data obtained by voice recognition performed on content being sung in real time by a performer may be given as rap lyric information in real time.
It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention cover modifications and variations that come within the scope of the appended claims and their equivalents. In particular, it is explicitly contemplated that any part or whole of any two or more of the embodiments and their modifications described above can be combined and regarded within the scope of the present invention.
Patent | Priority | Assignee | Title |
Patent | Priority | Assignee | Title |
10325581, | Sep 29 2017 | Yamaha Corporation | Singing voice edit assistant method and singing voice edit assistant device |
5621182, | Mar 23 1995 | Yamaha Corporation | Karaoke apparatus converting singing voice into model voice |
5703311, | Aug 03 1995 | Cisco Technology, Inc | Electronic musical apparatus for synthesizing vocal sounds using format sound synthesis techniques |
5747715, | Aug 04 1995 | Yamaha Corporation | Electronic musical apparatus using vocalized sounds to sing a song automatically |
5750912, | Jan 18 1996 | Yamaha Corporation | Formant converting apparatus modifying singing voice to emulate model voice |
5889223, | Mar 24 1997 | Yamaha Corporation | Karaoke apparatus converting gender of singing voice to match octave of song |
6337433, | Sep 24 1999 | Yamaha Corporation | Electronic musical instrument having performance guidance function, performance guidance method, and storage medium storing a program therefor |
6369311, | Jun 25 1999 | Yamaha Corporation | Apparatus and method for generating harmony tones based on given voice signal and performance data |
8008563, | Apr 12 2010 | Electronic circuit driven, inter-active, plural sensory stimuli apparatus and comprehensive method to teach, with no instructor present, beginners as young as two years old to play a piano/keyboard type musical instrument and to read and correctly respond to standard music notation for said instruments | |
20020005111, | |||
20020017187, | |||
20030009344, | |||
20040040434, | |||
20050137862, | |||
20050257667, | |||
20060015344, | |||
20060111908, | |||
20060173676, | |||
20090306987, | |||
20090307207, | |||
20110000360, | |||
20140006031, | |||
20160111083, | |||
20170025115, | |||
20170140745, | |||
20180018949, | |||
20180277075, | |||
20180277077, | |||
20180277080, | |||
20190096372, | |||
20190096373, | |||
20190096379, | |||
20190198001, | |||
20190304327, | |||
20190318712, | |||
20190318715, | |||
20190392798, | |||
20190392799, | |||
20190392807, | |||
20210012758, | |||
20210027753, | |||
20210193114, | |||
20210295819, | |||
20220076651, | |||
20220076658, | |||
EP2270773, | |||
EP2930714, | |||
EP3159892, | |||
JP200167078, | |||
JP200486067, | |||
JP2005331806, | |||
JP2006146095, | |||
JP2011013454, | |||
JP2013231872, | |||
JP201410190, | |||
JP201462969, | |||
JP2016206323, | |||
JP2017107228, | |||
JP2017194594, | |||
JP201727021, | |||
JP201797176, | |||
JP4238384, | |||
JP6332449, | |||
JP9050287, |
Executed on | Assignor | Assignee | Conveyance | Frame | Reel | Doc |
Feb 27 2020 | TACHIBANA, TOSHIYUKI | CASIO COMPUTER CO , LTD | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 052069 | /0809 | |
Mar 10 2020 | Casio Computer Co., Ltd. | (assignment on the face of the patent) | / |
Date | Maintenance Fee Events |
Mar 10 2020 | BIG: Entity status set to Undiscounted (note the period is included in the code). |
Date | Maintenance Schedule |
Aug 16 2025 | 4 years fee payment window open |
Feb 16 2026 | 6 months grace period start (w surcharge) |
Aug 16 2026 | patent expiry (for year 4) |
Aug 16 2028 | 2 years to revive unintentionally abandoned end. (for year 4) |
Aug 16 2029 | 8 years fee payment window open |
Feb 16 2030 | 6 months grace period start (w surcharge) |
Aug 16 2030 | patent expiry (for year 8) |
Aug 16 2032 | 2 years to revive unintentionally abandoned end. (for year 8) |
Aug 16 2033 | 12 years fee payment window open |
Feb 16 2034 | 6 months grace period start (w surcharge) |
Aug 16 2034 | patent expiry (for year 12) |
Aug 16 2036 | 2 years to revive unintentionally abandoned end. (for year 12) |